codengo

aka eric douglas brown

senior software engineer building intelligent systems in go. crafting neural networks from pure mathematics. four decades of code, one endless obsession.

40+

years coding

11+

years in go

neural architectures

01 — journey

the path

1981

the genesis

father builds a zenith heathkit h89. the first computer enters our home. a spark ignites in a child's mind.

1984

first lines

age seven. qbasic. msbasic. teaching myself in the green glow of a crt monitor. the obsession begins.

1990s

deep systems

c. c++. diving into memory management, pointers, the metal beneath the abstractions.

2013

go enters

discovered golang. simplicity meeting power. concurrency as a first-class citizen. home.

2016 - 2024

enterprise scale

building high-performance apis at top fortune 50 companies. millions of requests. zero excuses.

2024+

ai from scratch

building mlp, kan, esn neural networks in pure go. no frameworks. just mathematics and determination.

now

convergence

where decades of experience meet cutting-edge ai. senior engineer. systems architect. builder.

03 — expertise

technical depth

001

go / golang

primary language since 2013. high-performance systems, concurrent architectures, clean abstractions.

concurrencymicroservicescli

002

machine learning

neural networks from first principles. mlp, kan, esn architectures built without frameworks.

deep learningbackpropoptimization

003

systems programming

low-level optimization. understanding the machine at its core.

c/c++memoryperformance

004

backend architecture

high-performance go services. restful apis, grpc, middleware chains, clean architecture patterns.

restgrpcmiddleware

005

cloud & devops

container orchestration, ci/cd pipelines, cloud-native infrastructure.

dockerkubernetesaws

006

data systems

sql, nosql, data modeling, high-throughput architectures.

postgresqlredismongodb

04 — work

neural implementations

nn—001

multi-layer perceptron

pure go implementation. backpropagation, configurable layers, optimized matrix operations.

nn—002

kolmogorov-arnold network

learnable activation functions on edges. novel architecture exploration in go.

nn—003

echo state network

reservoir computing. dynamic temporal patterns, efficient training via linear regression.

                    // package neural implements feedforward neural networks with
                        configurable layers and backpropagation training.

                    package neural

                    type network struct {

                        layers    []layer

                        weights   [][][]float64

                        biases    [][]float64

                    }

                    func (n *network) forward(input
                    []float64) []float64 {

                        // no frameworks. just math.

                        return n.propagate(input)

                    }

codengo

the path

neural architecture

technical depth

neural implementations

let's build